All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: qemu-devel@nongnu.org
Cc: qemu-block@nongnu.org, kwolf@nongnu.org, jsnow@redhat.com,
	Stefan Hajnoczi <stefanha@redhat.com>,
	Fam Zheng <famz@redhat.com>, Kevin Wolf <kwolf@redhat.com>,
	Max Reitz <mreitz@redhat.com>
Subject: [Qemu-devel] [PATCH 07/31] block: Switch bdrv_co_get_block_status() to byte-based
Date: Mon, 17 Apr 2017 20:33:32 -0500	[thread overview]
Message-ID: <20170418013356.3578-8-eblake@redhat.com> (raw)
In-Reply-To: <20170418013356.3578-1-eblake@redhat.com>

We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based.  Convert another internal
function (no semantic change), and as with its public counterpart,
rename to bdrv_co_block_status() to make the compiler enforce that
we catch all uses.  For now, we assert that callers still pass
aligned data, but ultimately, this will be the function where we
hand off to a byte-based driver callback, and will eventually need
to add logic to ensure we round calls according to the driver's
request_alignment then touch up the result handed back to the
caller, if we start permitting a caller to pass unaligned offsets.

While at it, adjust the function to accept NULL for pnum or file,
while still guaranteeing the driver callback has a non-NULL pointer
to write into.

Signed-off-by: Eric Blake <eblake@redhat.com>
---
 block/io.c | 88 +++++++++++++++++++++++++++++++++++++-------------------------
 1 file changed, 53 insertions(+), 35 deletions(-)

diff --git a/block/io.c b/block/io.c
index 5cdb1f0..e804bdd 100644
--- a/block/io.c
+++ b/block/io.c
@@ -1696,69 +1696,83 @@ typedef struct BdrvCoGetBlockStatusData {
  * Drivers not implementing the functionality are assumed to not support
  * backing files, hence all their sectors are reported as allocated.
  *
- * If 'sector_num' is beyond the end of the disk image the return value is 0
+ * If 'offset' is beyond the end of the disk image the return value is 0
  * and 'pnum' is set to 0.
  *
- * 'pnum' is set to the number of sectors (including and immediately following
- * the specified sector) that are known to be in the same
- * allocated/unallocated state.
+ * 'pnum' is set to the number of bytes (including and immediately following
+ * the specified offset) that are known to be in the same
+ * allocated/unallocated state.  It may be NULL.
  *
- * 'nb_sectors' is the max value 'pnum' should be set to.  If nb_sectors goes
+ * 'bytes' is the max value 'pnum' should be set to.  If bytes goes
  * beyond the end of the disk image it will be clamped.
  *
- * If returned value is positive and BDRV_BLOCK_OFFSET_VALID bit is set, 'file'
- * points to the BDS which the sector range is allocated in.
+ * If returned value is positive and BDRV_BLOCK_OFFSET_VALID bit is set, and
+ * 'file' is not NULL, then *file points to the BDS which owns the allocated
+ * sector that contains offset.
  */
-static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
-                                                     int64_t sector_num,
-                                                     int nb_sectors, int *pnum,
-                                                     BlockDriverState **file)
+static int64_t coroutine_fn bdrv_co_block_status(BlockDriverState *bs,
+                                                 int64_t offset, int64_t bytes,
+                                                 int64_t *pnum,
+                                                 BlockDriverState **file)
 {
-    int64_t total_sectors;
-    int64_t n;
+    int64_t total_size;
+    int64_t n; /* bytes */
     int64_t ret, ret2;
+    int count; /* sectors */
+    BlockDriverState *tmp_file;

-    total_sectors = bdrv_nb_sectors(bs);
-    if (total_sectors < 0) {
-        return total_sectors;
+    total_size = bdrv_getlength(bs);
+    if (total_size < 0) {
+        return total_size;
     }

-    if (sector_num >= total_sectors) {
+    if (!pnum) {
+        pnum = &n;
+    }
+    if (offset >= total_size) {
         *pnum = 0;
         return 0;
     }

-    n = total_sectors - sector_num;
-    if (n < nb_sectors) {
-        nb_sectors = n;
+    n = total_size - offset;
+    if (n < bytes) {
+        bytes = n;
     }

     if (!bs->drv->bdrv_co_get_block_status) {
-        *pnum = nb_sectors;
+        *pnum = bytes;
         ret = BDRV_BLOCK_DATA | BDRV_BLOCK_ALLOCATED;
         if (bs->drv->protocol_name) {
-            ret |= BDRV_BLOCK_OFFSET_VALID | (sector_num * BDRV_SECTOR_SIZE);
+            ret |= BDRV_BLOCK_OFFSET_VALID | (offset & BDRV_BLOCK_OFFSET_MASK);
         }
         return ret;
     }

+    if (!file) {
+        file = &tmp_file;
+    }
     *file = NULL;
     bdrv_inc_in_flight(bs);
-    ret = bs->drv->bdrv_co_get_block_status(bs, sector_num, nb_sectors, pnum,
+    /* TODO: Rather than require aligned offsets, we could instead
+     * round to the driver's request_alignment here, then touch up
+     * count afterwards back to the caller's expectations.  But first
+     * we want to switch the driver callback to likewise be
+     * byte-based. */
+    assert(QEMU_IS_ALIGNED(offset | bytes, BDRV_SECTOR_SIZE));
+    ret = bs->drv->bdrv_co_get_block_status(bs, offset >> BDRV_SECTOR_BITS,
+                                            bytes >> BDRV_SECTOR_BITS, &count,
                                             file);
     if (ret < 0) {
         *pnum = 0;
         goto out;
     }
+    *pnum = count * BDRV_SECTOR_SIZE;

     if (ret & BDRV_BLOCK_RAW) {
-        int64_t bytes = *pnum * BDRV_SECTOR_SIZE;
-
         assert(ret & BDRV_BLOCK_OFFSET_VALID);
         ret = bdrv_block_status(*file, ret & BDRV_BLOCK_OFFSET_MASK,
-                                bytes, &bytes, file);
-        assert(QEMU_IS_ALIGNED(bytes, BDRV_SECTOR_SIZE));
-        *pnum = bytes >> BDRV_SECTOR_BITS;
+                                *pnum, pnum, file);
+        assert(QEMU_IS_ALIGNED(*pnum, BDRV_SECTOR_SIZE));
         goto out;
     }

@@ -1769,8 +1783,8 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
             ret |= BDRV_BLOCK_ZERO;
         } else if (bs->backing) {
             BlockDriverState *bs2 = bs->backing->bs;
-            int64_t nb_sectors2 = bdrv_nb_sectors(bs2);
-            if (nb_sectors2 >= 0 && sector_num >= nb_sectors2) {
+            int64_t size2 = bdrv_getlength(bs2);
+            if (size2 >= 0 && offset >= size2) {
                 ret |= BDRV_BLOCK_ZERO;
             }
         }
@@ -1779,11 +1793,10 @@ static int64_t coroutine_fn bdrv_co_get_block_status(BlockDriverState *bs,
     if (*file && *file != bs &&
         (ret & BDRV_BLOCK_DATA) && !(ret & BDRV_BLOCK_ZERO) &&
         (ret & BDRV_BLOCK_OFFSET_VALID)) {
-        BlockDriverState *file2;
-        int file_pnum;
+        int64_t file_pnum;

-        ret2 = bdrv_co_get_block_status(*file, ret >> BDRV_SECTOR_BITS,
-                                        *pnum, &file_pnum, &file2);
+        ret2 = bdrv_co_block_status(*file, ret & BDRV_BLOCK_OFFSET_MASK,
+                                    *pnum, &file_pnum, NULL);
         if (ret2 >= 0) {
             /* Ignore errors.  This is just providing extra information, it
              * is useful but not necessary.
@@ -1818,7 +1831,12 @@ static int64_t coroutine_fn bdrv_co_get_block_status_above(BlockDriverState *bs,

     assert(bs != base);
     for (p = bs; p != base; p = backing_bs(p)) {
-        ret = bdrv_co_get_block_status(p, sector_num, nb_sectors, pnum, file);
+        int64_t count;
+
+        ret = bdrv_co_block_status(p, sector_num * BDRV_SECTOR_SIZE,
+                                   nb_sectors * BDRV_SECTOR_SIZE, &count,
+                                   file);
+        *pnum = count >> BDRV_SECTOR_BITS;
         if (ret < 0 || ret & BDRV_BLOCK_ALLOCATED) {
             break;
         }
-- 
2.9.3

  parent reply	other threads:[~2017-04-18  1:34 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-18  1:33 [Qemu-devel] [PATCH 00/31] make bdrv_get_block_status byte-based Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 01/31] block: Drop unused bdrv_round_sectors_to_clusters() Eric Blake
2017-04-26 21:41   ` John Snow
2017-04-18  1:33 ` [Qemu-devel] [PATCH 02/31] block: Make bdrv_round_to_clusters() signature more useful Eric Blake
2017-04-26 21:41   ` John Snow
2017-04-26 21:47     ` Eric Blake
2017-04-26 21:54       ` John Snow
2017-04-18  1:33 ` [Qemu-devel] [PATCH 03/31] qcow2: Switch is_zero_sectors() to byte-based Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 04/31] block: Switch bdrv_make_zero() " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 05/31] qemu-img: Switch get_block_status() " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 06/31] block: Convert bdrv_get_block_status() to bytes Eric Blake
2017-04-18 21:34   ` [Qemu-devel] [PATCH v1.5 06.5/31] fixup! " Eric Blake
2017-04-18  1:33 ` Eric Blake [this message]
2017-04-18  1:33 ` [Qemu-devel] [PATCH 08/31] block: Switch BdrvCoGetBlockStatusData to byte-based Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 09/31] block: Switch bdrv_co_get_block_status_above() " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 10/31] block: Convert bdrv_get_block_status_above() to bytes Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 11/31] block: Add .bdrv_co_block_status() callback Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 12/31] commit: Switch to .bdrv_co_block_status() Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 13/31] file-posix: " Eric Blake
2017-04-18 20:56   ` Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 14/31] gluster: " Eric Blake
2017-04-19 13:26   ` [Qemu-devel] [Qemu-block] " Niels de Vos
2017-04-19 14:06   ` [Qemu-devel] " Prasanna Kalever
2017-04-19 14:08     ` Prasanna Kalever
2017-04-18  1:33 ` [Qemu-devel] [PATCH 15/31] iscsi: Switch cluster_sectors to byte-based Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 16/31] iscsi: Switch iscsi_allocmap_update() " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 17/31] iscsi: Switch to .bdrv_co_block_status() Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 18/31] mirror: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 19/31] null: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 20/31] parallels: " Eric Blake
2017-04-18 10:28   ` Denis V. Lunev
2017-04-18  1:33 ` [Qemu-devel] [PATCH 21/31] qcow: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 22/31] qcow2: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 23/31] qed: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 24/31] raw: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 25/31] sheepdog: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 26/31] vdi: Avoid bitrot of debugging code Eric Blake
2017-04-18  5:13   ` Stefan Weil
2017-05-13 20:35     ` Philippe Mathieu-Daudé
2017-04-18  1:33 ` [Qemu-devel] [PATCH 27/31] vdi: Switch to .bdrv_co_block_status() Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 28/31] vmdk: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 29/31] vpc: " Eric Blake
2017-07-13 12:55   ` Kevin Wolf
2017-07-13 13:52     ` Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 30/31] vvfat: " Eric Blake
2017-04-18  1:33 ` [Qemu-devel] [PATCH 31/31] block: Drop unused .bdrv_co_get_block_status() Eric Blake
2017-04-18  1:37 ` [Qemu-devel] [Qemu-block] [PATCH 00/31] make bdrv_get_block_status byte-based Eric Blake
2017-04-18 20:23 ` [Qemu-devel] " Eric Blake

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170418013356.3578-8-eblake@redhat.com \
    --to=eblake@redhat.com \
    --cc=famz@redhat.com \
    --cc=jsnow@redhat.com \
    --cc=kwolf@nongnu.org \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.